We are back from another fabulous Oracle OpenWorld, where the lawyers of the world sit on the edge of their seats hanging on Larry’s every word and Oracle’s competitors salivate over the free advertising they will get during the keynote. This year, in addition to the continued flood of information meant to showcase Oracles’s cloud computing leadership, the big announcement was the autonomous database cloud – the world’s first self-driving database. Somewhere in there is a Tesla joke!
The Autonomous-IT Paradigm
Actually, the self-driving database is pretty interesting when you dive into it, although a few Database Administrators (DBAs) that I spoke with at the event didn’t quite share my enthusiasm. The claim of anything ‘autonomous’ is sure to be met wth skepticism so I decided to take a deeper look at it.
The premise here is that the autonomous database is self-driving, more reliable and at a more cost-efficient price point versus Amazon’s Relational Database Service (RDS). We, here at Kaminario, applaud Oracle’s efforts as it develops and promotes the autonomous database. This compliments our own efforts as we build an analytics-driven, autonomous composable storage offering to tackle the challenges posed by unpredictable cloud-era workloads.
The key element of our upcoming product and Oracle’s autonomous database is the notion of dynamic composability, which Oracle calls instant elasticity. In essence, this gives the application (e.g., the database in Oracle’s case) the ability to dynamically and automatically provision and configure the resources it needs based on changing application requirements. This is exactly the promise we will deliver with our storage orchestration platform, Kaminario Flex. Flex will enable this same self-driving capability for our upcoming NVMe-based architecture, K2.N.
Better Business Agility through Automation and Orchestration
The benefit of Kaminario Flex and Oracle’s autonomous database are clear – better business agility through automation and orchestration. Much like the path in the automotive world from horse-drawn buggies to self-driving cars, this presents the next logical step in the evolution of the database from the early hierarchical models, such as IBM’s Information Management System (IMS), and in the data storage world from the Williams-Kilburn tube in 1947 to the NVMe devices beginning to appear on the market today.
Picture a typical rack of storage – multiple controllers and shelves physically located with each other. Any expansion of a single array might require another array to be moved to a different rack to free up space. This physical constraint is only necessitated by the legacy architecture which dictates the controller-drive shelf marriage. In tomorrow’s world (K2.N and Flex), this physical constraint is gone. Controller nodes and storage nodes can be located anywhere on the fabric and through orchestration can be logically combined into a virtual private array.
As the needs of an application change, the automation and orchestration layer will search the fabric for available or underutilized resources (IOPS and bandwidth at the controller level and capacity at the shelf level). These resources can be dynamically added to the existing array to expand it according to the application need. To use Oracle’s term, this brings instant elasticity to the infrastructure layer.
It is an exciting time in the datacenter. Cloud computing is impacting every architectural decision, whether it is public versus private cloud or private cloud versus siloed infrastructure. With innovation such as the Oracle Autonomous Database and Kaminario’s K2.N with Flex, the ability to dynamically compose and recompose your infrastructure based on the needs of the application is reality. The business agility it provides to the enterprise is the real gem and enables the realization of cloud economics in the enterprise datacenter. Thanks Larry! Not just for entertaining us during your keynotes, but for helping Kaminario advance the dynamic composability message to the world.